Wednesday, October 26, 2011

A Work In Progress Out of Africa Model Part I: Preface and Disclaimers

We Know A Lot About Prehistory And Much Of This Knowledge Is Recent

New information on the migration of hominins out of Africa and into Eurasia and beyond emerges every few weeks. There is a great deal of evidence, from archaeology, from current population genetics, from ancient DNA, from ancient climate data, from archaeozoology data, from linguistics and more that is pertinent to piecing together what happened, and evidence in the last few years has materially advanced our knowledge.

There Is Still A Lot We Don't Know About Hominin Prehistory

What we know is incomplete. The ancient DNA evidence is sparse and grows more thin as one looks further back in time and as one wants to know about more than mtDNA that is more easily recovered than non-replicating Y-DNA or autosomal markers. Some dates and classifications of specimens may be inaccurate. Skeletal remains are quite rare in the period prior to the Upper Paleolithic and not abundant in the Upper Paleolithic. Our records of material culture are particularly sparse in places like Southeast Asia when modern humans first arrived where we would very much like to have more detail. Founder effects for current relict populations and limited preservation of ancient DNA make it had to make strong inferences about the overall genetic makeup of ancient modern human populations from what we know. Admixed populations and a lack of solid data on prehistoric demographic parameters can confound efforts to determine effective population sizes of proto-populations. The enterprise of trying to date branches of the human genetic phylogeny from genetic variation arising from mutations have inherent and probably unresolvable amounts of uncertainty and are not particularly well calibrated to the extent that they can provide insight at all. Linguistic and cultural markers that have survived to be documented or observed today don't have enough time depth to tell us all that much about people in the Upper Paleolithic and earlier eras.

Hominin prehistory is a puzzle that is missing a great many pieces.

More Data Produces Increasing Returns By Strengthening Our Capacity To Make Reasonable Inferences

This uncertainty should not, however, take away from the wealth of information that we do have and from the emergent capacity that are growing body of knowledge permits us to infer from what we do know. As we gain new information we learn not just a specific fact, but also how particular kinds of processes, in general, work. For example, ancient DNA data from Europe didn't just tell us that a person with a particular haplogroup and material culture lives in a particular place at a particular time. The collection of data points also helped refine our knowledge about how stable population genetics are over time and what kind of factors from other lines of evidence appear to be associated with (or not associated with) population genetic change.

A Work In Progress Narrative Is A Suitable Way To Describe What We Know.

Summing up what we know in a "what we know from certainty" and what inferences can we make from that data point on a case by case basis, understates the richness of that data set we have with all of its deep, logical relationships. This approach would also rival reading a Pentagon-speak mission statement for clarity, precision and ease of reading.

Rather than going that route, what I propose below is a "work in progress" narrative. While stated in the positive, it isn't intended to be authoritative, definitive and final. But, it is intended to be consistent with what we know and among the more probable of the possible narratives given what we do know. To the extent that there is really no way to make any reasonable inferences at all, it is silent. At almost every point, there are reasonable arguments for somewhat different absolutely dated timelines, for material culture continuity to mask significant changes in population genetics, for material culture shifts to involve population genetic continuity, for the scale of admixture events or splitting populations to be off by orders of magnitude, and even for sequences of events to be out of order. Also, I have evaluated plausibility wholistically, rather than on a case by case basis, with an eye towards assumptions that make the overall narrative arc, rather than particular details, maximally accurate.

This is also a work in progress in the sense that conclusions are frequently based sources not referenced in this first draft and that some pieces of the puzzle that the scholarly literature elucidates reasonably well are omitted in this first draft.


This narrative also intentionally not comprehensive. It omits almost all of the evolution of hominins from the African Great Apes within Africa prior to the emergence of hominins outside Africa. It discusses pre-modern human hominins, in general, only to the extent that detail about them is necessary to tell the story of modern humans and isn't particularly interested in what their lives were like or what impact they had for their own sake. It omits almost everything about the Holocene except to the extent that it helps to link the current population genetic landscape to the Upper Paleolithic population genetic landscape.

It omits a great deal of interregional detail within Africa. This narrative is agnostic on the point in time at which hominins other than modern humans went extinct in Africa. Some genetic evidence suggests that there could have been admixture between modern humans (mostly Pygmy and Khoisan) and archaic humans in Africa (probably two separate archaic human populations, at least) as recently as sometime in the last 20,000 years, but absence of corroborating material culture or skeletal remains for archaic humans until a much more remote point in the past, and the limited reliability of relatively untested and uncalibrated genome based dating protocols make the most recent result, which has not even been independently confirmed, calls for skepticism regarding this result at this stage, although it should not be dismissed out of hand.

No comments: